HAMi vs Moab HPC Suite
HAMi
project-hami.ioMoab HPC Suite
adaptivecomputing.comSide by side
| Figure | HAMi | Moab HPC Suite |
|---|---|---|
| Price from | Free | — |
| Free plan | Yes | — |
| Free trial | — | — |
| Open source | — | — |
| Linux | Yes | Yes |
| API | — | Yes |
| Self-hosted | Yes | Yes |
| Plans published | 1 | 1 |
| Spec score · GPU Cluster Management Software | 8.3 (#5) | 7.2 (#14) |
| Deployment | The quick start installs HAMi with Helm and requires Kubernetes, Helm, kubectl, and installation permissions. | The suite documentation describes manual and RPM installation approaches and recommends testing upgrades in an environment that mirrors production. |
| Purpose | HAMi is open-source, cloud-native GPU virtualization middleware for sharing, isolating, and scheduling heterogeneous accelerators on Kubernetes. | Moab HPC Suite optimizes resource utilization and job scheduling across high-performance computing infrastructure. |
| Scheduling | HAMi offers binpack, spread, and topology-aware scheduling policies. | Its scheduling algorithms optimize job placement and resource allocation to improve cluster utilization and reduce wait times. |
| Support | The site links to documentation, tutorials, Discord and the `#hami-dev` Slack channel for community resources. | Customers with a support services contract can submit unresolved support cases by email for help from a technical support specialist. |
Plans
In their own words
HAMi
HAMi is open-source middleware for sharing and scheduling GPUs and other accelerators among AI workloads on Kubernetes. Workloads can request GPU memory and core limits through Kubernetes resource controls. Scheduling options include binpack, spread and numa-first, with binpack and spread available at node or GPU scope. The supported-device list spans NVIDIA, AMD, AWS Neuron, Huawei Ascend, Cambricon, Enflame, Hygon, Iluvatar, Kunlunxin, MetaX, Moore Threads, Vastai and Biren; isolation and partitioning capabilities vary by device. HAMi works with Kubernetes APIs, DRA and CDI, and documents scheduling integrations including Volcano, Kueue, Koordinator and NVIDIA KAI Scheduler. Its node metrics endpoint exposes device memory and core utilization for Prometheus. Classic HAMi requires Kubernetes v1.23 or later. HAMi-DRA requires Kubernetes v1.34 or later, the DRA Consumable Capacity feature gate, CDI and NVIDIA driver 440 or later. NVIDIA limits use user-space library interception, which does not cover applications that bypass CUDA. Enforcement is described as soft and best-effort; the FAQ recommends MIG when hardware-enforced isolation is required.
Moab HPC Suite
Moab HPC Suite is self-hosted software for scheduling workloads and managing resources across high-performance computing infrastructure. It runs on Linux and exposes an API. Scheduling algorithms place jobs and assign resources to improve cluster utilization and reduce waiting. Automatic workload balancing and priority management direct capacity toward critical jobs. Administrators can define scheduling, allocation, and user-access policies, including quota controls. The suite provides cluster-performance and resource-use reporting, GPU utilization metrics, and management of multiple HPC clusters through one interface with unified monitoring and control. Its high-availability architecture includes failover capabilities for continuous operation of critical workloads. Enterprise Edition components include Moab Workload Manager, Moab Web Services, Moab Accounting Manager, and TORQUE Resource Manager. Moab Web Services provides REST access to jobs, nodes, virtual machines, and reservations. Documentation lists TORQUE 5.0.1 and SLURM support, manual and RPM installation, and guides for Red Hat, SUSE, and Ubuntu-based systems. Pricing is available on request.